Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We explore the hypothesis that traditional joint-liability microfinance programs fail to increase borrower incomes in part because they cannot screen out unproductive borrowers. In randomly selected villages in West Bengal, India, we implemented trader-agent-intermediated lending (TRAIL), in which local trader-lender agents were incentivized through repayment-based commissions to select borrowers for individual liability loans. In other randomly selected villages, we organized a group-based lending (GBL) program in which individuals formed 5-member groups and received joint liability loans. TRAIL loans increased the production of the leading cash crop by 27% and farm incomes by 22%. GBL loans had insignificant effects. We develop and test a theoretical model of borrower selection and incentives. Farmers selected by the TRAIL agents were more able than those who self-selected into the GBL scheme; this pattern of selection explains at least 30–40% of the observed difference in income impacts.